doi: 10.1016/j.neuron.2021.01.007. Epub 2021 Jan 28.
Interneuronal exchange and functional integration of synaptobrevin via extracellular vesicles
Affiliations
- PMID:33513363
- PMCID: PMC7979516
- DOI: 10.1016/j.neuron.2021.01.007
Item in Clipboard
Interneuronal exchange and functional integration of synaptobrevin via extracellular vesicles
A Alejandro Vilcaes et al. Neuron..
Display options
Format
Display options
Format
Abstract
Recent studies have investigated the composition and functional effects of extracellular vesicles (EVs) secreted by a variety of cell types. However, the mechanisms underlying the impact of these vesicles on neurotransmission remain unclear. Here, we isolated EVs secreted by rat and mouse hippocampal neurons and found that they contain synaptic-vesicle-associated proteins, in particular the vesicular SNARE (soluble N-ethylmaleimide-sensitive factor [NSF]-attachment protein receptor) synaptobrevin (also called VAMP). Using a combination of electrophysiology and live-fluorescence imaging, we demonstrate that this extracellular pool of synaptobrevins can rapidly integrate into the synaptic vesicle cycle of host neurons via a CD81-dependent process and selectively augment inhibitory neurotransmission as well as specifically rescue neurotransmission in synapses deficient in synaptobrevin. These findings uncover a novel means of interneuronal communication and functional coupling via exchange of vesicular SNAREs.
Keywords: CD81; extracellular vesicles; neurotransmission; synaptic vesicle; synaptobrevin-2.
Copyright © 2021 Elsevier Inc. All rights reserved.
Conflict of interest statement
Declaration of interests The authors declare no competing interests.
Figures
(A) Schematic representation of the EV isolation methods. First, a shared clearing procedure to remove sedimented live cells, dead cells, debris and microvesicles was performed. After filtering the supernatant through a 0.45 μm syringe filter, EVs were isolated using ultracentrifugation and two commercial kits following the manufacturer’s instructions. (B) Representative particle size distribution graphs obtained by nanoparticle tracking analysis (NTA) measurements of EVs isolated by the 3 different methods. Insets: morphologic observation of the EVs by transmission electron microscopy. Scale bar = 100 nm. (C) Classification of the proteome of EVs based on gene ontology and biological function/pathway annotations from Protein Information Resource (http://proteininformationresource.org ) and Panther Classification System (http://pantherdb.org ).
(A) Top: Representative confocal images of cultured hippocampal neurons immunofluorescently labeled with anti-synapsin 1 (red), anti-Tau (blue) and anti-CD81 (green). White scale bar = 2 μm. Bottom: fluorescence intensity quantification along the region of interest (ROI) depicted in the image. (B) Representative Western blot and quantification of CD81 protein levels in lysates from neurons infected with L-307 (Control) or CD81 KD lentiviruses. CD81 augments spontaneous inhibitory neurotransmission. (C) Representative traces, (D) average frequency and average amplitude (E) of mIPSCs in control neurons and infected with CD81, E219Q, E219A or CD81 KD lentivirus. Data was analyzed with one-way ANOVA (mIPSC frequency: F=17.55 p<0.0001 - results of the post hoc Dunnet’s test are represented in the figure -; mIPSC amplitude: F=1.449 p=0.2075). Inset in D: mIPSC frequency for non-infected (control) neurons and infected with the empty vectors pFUGW and L307. One-way ANOVA: F=0.4016 p=0.6703.
(A) Schematic representation of the experimental strategy. (B) Representative traces and (C) time course of frequency of mIPSCs in neurons incubated or not (Control) with EVs isolated by the three methods described in Fig. 1. Data was analyzed with a two-way ANOVA (time factor: F=6.906, p<0.0001; experimental group factor: F=17.94, p<0.0001). Multiple comparisons by Sidak’s post hoc test revealed p<0.0001 (****) for all three methods vs control, and p>0.9999 (non-significant - NS -) for EVs CD81 KD vs control (N is 5–10 neurons per time point per group). (D) Average amplitudes of mIPSCs in the conditions mentioned above (two-way ANOVA: F=0.4665 p=0.9749). EV-mediated augmentation of spontaneous inhibitory neurotransmission is activity-independent and calcium-dependent. (E) Representative traces, (F) average frequency and average amplitude (G) of mIPSCs in neurons untreated (control) or treated with EVs in the presence (spontaneous) or absence (basal activity) of TTX. Alternatively, control or EVs groups were pretreated along with the calcium chelator BAPTA-AM (100 μM). Data was analyzed using ANOVA (mIPSC frequency: two-way ANOVA, Group factor: F(4,83)=33.54 p<0.0001, Time factor: F(1,83)=2.272 p=0.1355 - results of the post hoc Sidak’s multiple comparisons test are represented in the figure –; mIPSC amplitude: one-way ANOVA, F=1.210 p=0.3124). Astrocyte-derived EVs do not regulate mIPSC. (H) Neurons were preincubated in TTX alone (control) or plus EVs from neuronal or astrocytic origin. Left: Frequency of mIPSC. Right: Amplitude of mIPSC. Data was analyzed with ANOVA (mIPSC frequency: two-way ANOVA, Group factor: F(2,64)=31.88 p<0.0001, Time factor: F(1,64)=0.5022 p=0.4811; mIPSC amplitude: one-way ANOVA, F=0.5961 p=0.5539).
(A) Schematic representation of the experimental strategy. (B) Immunofluorescence (IF) analysis of subcellular localization of GFP (to label syb2-pHluorin) and syn1. White scale bars: 10 μm for top image, 1 μm for the rest. Line scans showing similar spatial distribution for GFP and syn1 are shown. Left bar graph: quantification of the number of GFP (pHluorin) positive cell bodies relative to total neurons per image field. Right bar graph: object-based colocalization analysis and quantification of syn1 boutons containing GFP signal. (C) Live fluorescence imaging of syb2-pHluorin incorporated from EVs. Left: representative images of putative presynaptic boutons before and after perfusion of 50 mM NH4Cl. Right: fluorescence traces of the two regions of interests (ROI) depicted in the images. White scale bar = 1 μm. (D) Representative fluorescence images of exogenous syb2-pHluorin incorporated from EVs before (top) and after (bottom) stimulation at 40 Hz for 5 s (200 AP). White scale bar = 2 μm. (E) Example fluorescence traces of putative presynaptic terminals (active boutons, left) and spots of background fluorescence (top center). Center bottom: Average fluorescence traces from 4 independent experiments (thin lines; >15 recordings per experiment and >5 ROIs per recording) and ensemble average (thick lines). Inset (top right): Histogram of fluorescence amplitudes (normalized to F0) and lognormal fit (blue line). EVs isolated from CD81 KD neurons have reduced incorporation into target WT neurons. (F) Representative immunofluorescence images and (G) object-based colocalization analysis and quantification of syn1 boutons containing GFP signal after incubation of WT neurons with EVs isolated from CD81 KD + syb2-pHluorin neuron cultures.
(A) Representative traces and (B) time course of frequency of mIPSCs in Syb2 KO neurons incubated (Syb2 KO + EVs) or not with EVs. Two-way ANOVA: time effect F=2.575 p=0.0400, experimental group effect F=18.69 p<0.0001 (N is 4–6 neurons per time point per group). Inset: Average mIPSC frequency at 50–60 min comparing wild type (control) neurons with syb2 KO with and without EVs (syb2 KO vs syb2 KO + EVs p=0.0075 by unpaired t-test). (C) Average amplitude of mIPSCs in the conditions mentioned above (syb2 KO vs syb2 KO + EVs p=0.8775 by unpaired t-test). (D) Frequency and (E) Amplitude of mEPSC in Syb2 KO hippocampal neurons at 30–60 min after addition of EVs (unpaired t-test: frequency p=0.0471, amplitude p=0.9863). CD81 KD impairs Syb2 recruitment and secretion via EVs. (F) Dot-blot of EVs isolated from different sources, immunostained against Syb2 (whole brain lysates were used as positive controls). EVs do not increase evoked release in Syb2 KO neurons. (G) Representative eIPSC traces from syb2 KO neurons incubated with or without EVs. (H) Time course of eIPSC amplitude after EVs addition. Two-way ANOVA: time factor F=0.9299 p=0.4474, experimental group factor F=43.79 p<0.0001 (post hoc Tukey’s test revealed p<0.0001 for control vs syb2 KO and sun2 KO + EVs, p=0.7112 - NS - for syb2 KO vs syb2 KO + EVs). EVs that lack Syb2 do not modulate neurotransmission. (I) Dot-blot of EVs isolated from Syb2 KO neurons and their littermate controls, Syb2 heterozygous (Het; positive control = whole brain lysate). (J) Representative traces, (K–L) Frequency and (M) Amplitude of mIPSC in Syb2 KO and WT mouse hippocampal neurons incubated for 30–60 min with EVs isolated from Syb2 KO or Het neuron cultures. Data was analyzed with one-way ANOVA (mIPSC frequency in Syb2 KO: F=6.388 p=0.0036; mIPSC frequency in WT: unpaired t-test p=0.5356; mIPSC amplitude: F=0.9418 p=0.4484).
(A) Representative traces and (B) time course of frequency of mIPSCs in SNAP25 KO neurons incubated (SNAP25 KO + EVs) or not with EVs. Two-way ANOVA: time effect F=1.304 p=0.3017, experimental group effect F=0.02381 p=0.8789 (N is 3–4 neurons per time point per group). (C) Average amplitude of mIPSCs in the conditions mentioned above (SNAP25 KO vs SNAP25 KO + EVs p=0.8900 by unpaired t-test). (D). Top: Schematic representation of the experiment. Center: Average fluorescence traces in SNAP25 KO neurons and littermate control - WT - (2 independent experiments, 8–12 recordings per experiment and >5 ROIs per recording). Bottom: Fluorescence amplitudes (normalized to F0). (E) Representative immunofluorescence images of SNAP25 KO neurons incubated with Syb2-pHluorin EVs and immunostained against GFP, Syn1 and Tau. (F) Object-based colocalization analysis and quantification of fraction of GFP-positive (syb2-pHluorin) bouton-like objects colocalizing to presynaptic terminals (syn1). (G) Object-based colocalization analysis and quantification of Syn1 boutons containing GFP signal after incubation of SNAP25 KO neurons with EVs syb2-pHluorin neuron cultures. (H) Schematic representation of the mechanism of EVs-mediated augmentation of spontaneous neurotransmission: CD81 participates in the recruitment of Syb2 to EVs for posterior secretion. EVs are then incorporated by the target neuron and the exogenous Syb2 can be functionally integrated into synaptic vesicles increasing their propensity to release.
References
- Bachurski D, Schuldner M, Nguyen PH, Malz A, Reiners KS, Grenzi PC, Babatz F, Schauss AC, Hansen HP, Hallek M, and Pogge von Strandmann E (2019). Extracellular vesicle measurements with nanoparticle tracking analysis - An accuracy and repeatability comparison between NanoSight NS300 and ZetaView. J Extracell Vesicles 8, 1596016. - PMC - PubMed
- Beaudoin GM, Lee SH, Singh D, Yuan Y, Ng YG, Reichardt LF, and Arikkath J (2012). Culturing pyramidal neurons from the early postnatal mouse hippocampus and cortex. Nat Protoc 7, 1741–1754. - PubMed
Publication types
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases
Research Materials